How to get upwelling TOA radiance in visible/swir?

[carolinecox]

I would like to simulate TOA total upwelling radiances in the visible/SWIR region. Having looked at the LBLRTM instructions, and online, I think what I have to do is run two TAPE5's, and I have copied what I have below.

• generate solar file first - I can do this ok

My first TAPE 5:

$ Run 1 TAPE5, produces DW radiance and transmittance, set angle to sza. Copy the TAPE12 to SOL.PATH.T2 HI=1 F4=1 CN=1 AE=0 EM=1 SC=0 FI=0 PL=0 TS=0 AM=1 MG=0 LA=0 OD=0 XS=0 00 00 20513.00 21622.00 1.0 0.0 0.0 0.0000 0.000 0 0.0 0.0000 s 2 2 0 0 0
0.00 100.00 0.00

-1. %

My second TAPE5 $ Run 2 TAPE5, produces UW radiance and transmittance, set angle to vza.
HI=1 F4=1 CN=1 AE=0 EM=1 SC=0 FI=0 PL=0 TS=0 AM=1 MG=0 LA=0 OD=0 XS=0 00 00 20513.00 21622.00 1.0 0.0 0.0 0.0000 0.000 0 0.0 294.2000 -1 2 2 0 0 0
100.00 0.00 180.00

-1. $ Convolve transmittance with solar source, needs SOL.PATH.T2 and TAPE12 as input HI=0 F4=0 CN=0 AE=0 EM=2 SC=0 FI=0 PL=0 TS=0 AM=0 MG=0 LA=0 MS=0 XS=0 0 0 2 2 2 -1. $ Transfer to ASCII plotting data (TAPES 27 ) HI=0 F4=0 CN=0 AE=0 EM=0 SC=0 FI=0 PL=1 TS=0 AM=0 MG=0 LA=0 MS=0 XS=0 0 0

Plot title not used

20513.00 21622.00 10.2000 100.0000 5 0 13 0 1.000 0 0 0
0.0000 1.2000 7.0200 0.2000 4 0 1 1 0 0 0 3 27 -1. %

Is this the correct procedure? It would be really handy to have some more examples to look at for running LBLRTM in different configurations, Many thanks Caroline

[carolinecox]

Hi, further to this comment, I have been looking into the code, and I see

! ------------------------------------------------------------ ! NOTE: REFLECTION FUNCTION NOT CURRENTLY INCORPORATED ! INTO SOLAR RADIATIVE TRANSFER CALCULATION

So, does this mean the surface reflectivity that I input is not used?? That seems to be the case looking at my simulation.

[kcadyper]

solar_test_2023.tar.gz HI, Caroline

Thanks for your interest in LBLRTM. Before I reply to your questions a quick comment: please be aware that LBLRTM does not model scattering, except for Rayleigh extinction.

You can model surface reflection; unfortunately the process is a bit clunky. LBLRTM is a very old code, and some sections have not been updated in many years. Rather than directly using the ASCII REFLECTIVITY file (which I think you are trying to do) you need to convert it to binary. To do so you need to compile and run a FORTRAN code (in the zip file I have attached) called wrtlblbinary, which creates a file called SOL.REFLECTANCE. The zip file also contains sample TAPE5s for each step of the process and a Perl script to run the three steps. You may not want to run the script but it has comments that describe each step. Based on your TAPE5s I think you have a good understanding of the setup, but there are some details tripping you up. First, the TAPE5s are read in using FORTRAN fixed format, which means that the inputs need to be in specific places on each line. It may be that you did this and GitHub just collapsed the spaces, in which case I apologize. The example TAPE5s show you where each input has to go. Second, in the upwelling TAPE5 you need a "-1" in both the emissivity and reflectivity fields, unless you are using a zero value for the surface emissivity. Finally, in the TAPE5 for the run applying the solar source, you selected a Julian date of 2. Was this deliberate? Nothing wrong with this, just curious.

Let me know if you have further questions.

Karen